Seasonal regression of brain regions involved in the control of birdsong provides a striking and unique opportunity to investigate the mechanisms regulating neuronal degeneration and protection associated with naturally occurring variation in steroid hormones, and the functional consequences of neuroprotection for a learned sensorimotor behavior. The aims in this proposal address fundamental issues of hormones as neuroprotective agents in adult brains. These include the role of indirect genomic signaling pathways in hormonal neuroprotection (Aims 1-4), the role of kinase cascades in mediating transynaptic neuroprotective effects of hormones (Aim 3), and whether steroids can have neuroprotective effects in a manner independent of hormone receptors (Aim 2). The birdsong system excels as a model for studies of hormonal mechanisms of neuroprotection. It is a well-defined and tractable neural circuit that shows extreme seasonal patterns of hormone-regulated neuronal regression and protection. These processes of neural degeneration and protection occur with breeding-related hormonal cycles and thus can be studied in vivo without invasive manipulations. This research will advance the field by 1) providing the first evidence in the song system that kinase cascades are indirect genomic contributors to hormonal neuroprotection (Aims 1-4); 2) investigating the mechanisms by which hormones act transynaptically to have neuroprotective effects, and whether kinase cascades mediate this effect (Aim 3); and 3) determining whether the contribution of kinase cascades to the neuroprotective effect of steroids requires hormone receptor activation (Aim 2), an issue of continuing uncertainty.